Software Architecture
The software architecture of BMB (Bionic Mosquito Bot) is engineered to support fully autonomous operation, real-time decision-making, adaptive mission logic, and resilient swarm communication—all within the constraints of a micro-scale airborne platform. Inspired by the layered approach found in robust embedded systems, BMB's software is divided into well-defined tiers, each responsible for a critical aspect of flight, sensing, behavior, or communication.
Firmware Layer
At the foundation of BMB's intelligence lies the firmware layer, responsible for low-level, deterministic control of the bot's physical systems. This ultra-lightweight codebase handles:
PID-Based Flight Control: Custom-tuned control loops regulate the flapping amplitude, frequency, and wing phase offset for stable lift and directional control.
Sensor Synchronization: The firmware orchestrates sensor polling cycles to ensure synchronized multi-modal data capture, critical for coherent SLAM and behavioral response.
Power Management: Microsecond-resolution timers monitor current draw, battery status, and thermal readings, enabling real-time load balancing and triggering emergency power-saving states if thresholds are breached.
Emergency Protocols: Collision detection triggers instant auto-hover or soft landing routines using inertial sensors and gyros. If loss of signal or catastrophic system error occurs, BMB reverts to autonomous return-to-base mode using its last-known navigation cache.
All firmware routines are interrupt-driven and optimized for latency-sensitive operation. Updates to the firmware layer are validated via cryptographic checksums and failover rollback mechanisms.
Micro OS Layer
Above the firmware is a minimalist, microkernel-based real-time operating system (RTOS) designed specifically for ultra-constrained environments. Key features include:
Task Isolation: Each module—flight control, sensor fusion, AI inference—runs in its own isolated thread with strict memory partitioning, reducing fault propagation.
Event Scheduler: A low-jitter scheduler manages real-time and best-effort processes, ensuring stable performance even under high computational loads.
Diagnostics & Telemetry Logging: Runtime status, fault codes, and mission performance metrics are logged to encrypted flash and can be transmitted during burst sync cycles.
OTA Update Support: Secure, fail-safe over-the-air updates allow remote patching or upgrading of system components without interrupting operational integrity.
The Micro OS also exposes standard APIs to higher-level behavioral logic, supporting modular development and future extensibility.
Navigation & Behavior Layer
This layer comprises the brain behind BMB’s autonomy and mission execution. It translates abstract mission goals into context-aware actions, balancing mobility, safety, and data collection. Core components include:
SLAM & 3D Mapping Engine: Uses Lidar and IR data fused with IMU inputs to build real-time volumetric maps of the surrounding space. Supports dynamic re-mapping for unknown or changing environments.
Flight Strategy Module: Chooses between behaviors such as patrol, hover-inspection, exploration, or tracking based on sensor input and mission configuration.
Flock AI Module: Implements distributed swarm intelligence, enabling BMB units to maintain formation, share local maps, avoid mid-air collisions, and cooperatively explore larger spaces.
Recognition & Detection Routines: Lightweight convolutional neural networks (CNNs) classify observed patterns, such as:
Human silhouettes or facial heat signatures
Moving targets (e.g., intruders, wildlife)
Hazardous areas (e.g., fire zones, chemical plumes)
These models are dynamically loaded into memory depending on the mission type, ensuring efficient resource utilization.
Communication Layer
Communication in BMB is designed to be secure, adaptive, and fault-tolerant, leveraging a hybrid of short-range and medium-range wireless protocols:
Mesh Radio Protocol: A decentralized, peer-to-peer mesh protocol allows real-time swarm coordination, data relaying, and task delegation between BMB units.
Burst Transmission: BMB can sync high-resolution data or full mission logs via Bluetooth Low Energy (BLE) or Ultra-Wideband (UWB) to nearby edge gateways or ground stations.
WiFi Module (Optional): For high-throughput scenarios, the bot can initiate short-range WiFi transfers when docked, uploading HD imagery, sensor archives, and diagnostics.
Security Protocols: All communication is secured using AES-256 encryption with rotating session keys. Mutual authentication protocols prevent spoofing, man-in-the-middle attacks, or unauthorized control attempts.
In disconnected or signal-denied environments, BMB queues critical data for later upload and autonomously switches to fail-safe communication fallback modes.
This layered software architecture allows BMB to operate intelligently and independently across a wide range of mission profiles. Its real-time autonomy, swarm intelligence, and secure communication make it a reliable agent for complex micro-sensing tasks in both civil and tactical deployments.
Last updated